If you use Chrome or Edge, you already have a way to listen to a web page for free. So it’s fair to ask what an extension like Spoken adds. This post compares Spoken with three free options as of November 2026: Read aloud in Chrome’s Reading mode, Read Aloud in Microsoft Edge, and the open-source Read Aloud extension. Each one is good at something, and we try to say where.
The short version
| Chrome Reading mode | Edge Read Aloud | Read Aloud extension | Spoken | |
|---|---|---|---|---|
| Install | Built in | Built in | Extension | Extension |
| Voices | Google natural voices, system voice fallback | Natural voices, some online, a few offline | System voices, plus optional cloud voices | 21 Kokoro voices |
| Where speech is made | Voices download to your device for your language | On the device or on Azure, depending on the voice | On the device for system voices, on a cloud server for premium voices | In your browser, on WebGPU or the CPU |
| Languages | Per language you add | Many voices and accents, varies by market | 40+ (per its store listing) | English only (US and UK) |
| Highlighting | Adaptive, word, sentence or off | Not stated on Microsoft’s Read aloud page | Yes, set in Options | Sentence, plus an estimated word, five colours |
| Start from a paragraph | Previous and next sentence | Not covered here | Not covered here | A marker beside each paragraph |
| PDFs | Not stated in Google’s help | Yes (announced May 2020) | Yes | Yes, in Chrome and Edge’s PDF viewer |
| Platforms | Desktop and Chromebook (Reading mode); Android (Listen to this page) | Desktop, iOS and Android | Chrome, Firefox and Edge | Desktop Chrome and Edge (version 125+), no mobile |
A “Not covered here” or “Not stated” cell means we found no official source for that feature, not that the feature is missing. On Android, Chrome also has Listen to this page, which reads pages in 25 languages, can highlight text and auto scroll, and keeps playing when the screen is locked, according to Google’s help page.
Voice quality
Chrome downloads Google natural voices for your current language the first time you open Reading mode. If the download fails, it uses your system voice instead. You can preview voices and pick one.
Edge offers what Microsoft calls natural-sounding voices in a range of accents. Its online voices are generated by Azure Cognitive Services on Microsoft’s servers, according to Microsoft’s privacy whitepaper. Microsoft notes that availability varies by device, market and browser version.
The Read Aloud extension uses your browser’s system voices by default. It can also use cloud voices from Google Wavenet, Amazon Polly, IBM Watson, Microsoft Azure and OpenAI. Some of these need an in-app purchase or your own API key.
Spoken runs one model, Kokoro-82M, with 21 English voices: 11 American female, 8 American male, one British female and one British male. You can preview each one before choosing it.
To be plain about it: large cloud voices may sound more expressive than a small model running on your laptop, and they cover far more languages. Spoken is English only today. Spoken doesn’t yet have its own text preprocessing beyond the voice model’s basic normalizer, so URLs, long numbers and acronyms can come out awkwardly. Voice quality is a matter of ear, so the best test is to play the same paragraph in each and decide for yourself.
Where your text goes
This is where the options differ most, and it depends on the voice you pick, not just the product.
Edge is the clearest about it. Microsoft’s privacy whitepaper says text is converted either by a client-side library from the platform or OS, or by Azure on Microsoft’s servers. Voices with “Online” in the name send the text and a random token to Microsoft, which returns audio. Microsoft says the data is encrypted in transit and deleted right after conversion. Client-side voices send nothing to Microsoft. Only a few voices work offline.
Chrome downloads its natural voices to your device. Google’s Reading mode help page doesn’t go further on where text is processed.
The Read Aloud extension’s privacy policy says local voices synthesize speech on your computer and never send text to the internet. Premium voices send the text to its cloud server, which may cache it briefly. Standard voices need no account; premium voices ask you to sign in with Google.
Spoken makes speech inside your browser, on your graphics card through WebGPU, or on your processor if WebGPU isn’t available. Spoken makes no network requests with your text. Its one third-party request is the model download from Hugging Face: about 326 MB for the WebGPU model or about 92 MB for the processor fallback. When you read a PDF, it also fetches that PDF again from the site it came from. After the download the model stays in the browser’s cache, unless you clear site data. There is no analytics or telemetry. The engine panel says which mode is running and “Nothing leaves this browser.”
The trade-off is real. Cloud voices don’t use your GPU, CPU or battery and work on low-end hardware. Spoken needs that first download and a reasonably capable machine.
On the page
Chrome reads inside Reading mode, which strips page clutter into a cleaner reading view. You get play and pause, previous and next sentence, speed, and highlighting at adaptive, word or sentence level. Reading mode doesn’t work on every site; Google lists video, forum, gallery, map, form, login and payment pages among the exceptions.
Edge reads from the menu or from a right-click “Read aloud selection”.
Spoken works on the page itself rather than in a separate view:
- Highlighting. The current sentence is highlighted in place without changing the page’s markup. The current word is highlighted too, but its position is estimated from the audio timing, so it can run slightly ahead of or behind the voice. You can track the word, the sentence or both, in violet, amber, mint, rose or sky.
- Read from here. On a detected article, a small orb sits beside each paragraph. Click one to read from that paragraph to the end.
- Listen to this article. Spoken can offer a pill with the reading time. Nothing plays until you press Start.
- Selection. Select text and an orb pill offers to speak it, or use the right-click “Speak selection” menu, or press Alt+Shift+S (Alt+Shift+K on ChromeOS).
- The orb. The play button is an orb that reacts to the voice as it speaks. A docked player shows your place. For a selection it reads “Sentence n of N” and has speed, play/pause and stop buttons. For an article it reads “Paragraph n of N” and has a stop button.
There is no seek bar yet. Paragraph orbs are the way to jump around.
PDFs
Edge has read PDFs aloud in its built-in viewer since Microsoft announced it in May 2020. The Read Aloud extension reads PDFs, Google Docs, Google Play Books and Kindle. Chrome’s help pages don’t say whether Read aloud works on PDFs, and we couldn’t confirm it either way.
Spoken reads PDFs in the browser’s own viewer, in both Chrome and Edge. Text is extracted locally with a bundled copy of pdf.js, and you get the same selection reading, highlighting, paragraph orbs and a “Listen to this PDF” pill. PDFs up to 100 MiB are supported, and you can turn PDF reading off in settings.
Which one to use
If you want the most languages, a phone (Edge on iOS and Android, Chrome’s Listen to this page on Android), or no install at all, Chrome and Edge’s built-in options are good and free. Edge in particular publishes a clear account of which voices stay local. If you want lots of voice providers, there is the Read Aloud extension, which is open source and lists about 6 million users on the Chrome Web Store as of November 2026.
If you read mostly in English, want neural voices that run entirely in your browser, and want to listen without leaving the page, Spoken is built for that. Give it a try on the next article you were going to skim.
Sources (checked 2026-11-14)
- Chrome Reading mode and Read aloud: https://support.google.com/chrome/answer/14218344
- Chromebook Reading mode voices: https://support.google.com/chromebook/answer/15284142?hl=en
- Chrome on Android, Listen to this page: https://support.google.com/chrome/answer/14768725
- Microsoft Edge Read aloud: https://explore.microsoft.com/en-us/edge/features/read-aloud
- Microsoft Edge privacy whitepaper: https://learn.microsoft.com/microsoft-edge/privacy-whitepaper
- Edge Read aloud PDF support: https://techcommunity.microsoft.com/t5/articles/microsoft-edge-read-aloud-now-with-pdf-support/m-p/1951068
- Read Aloud on the Chrome Web Store: https://chromewebstore.google.com/detail/read-aloud-a-text-to-spee/hdhinadidafjejdhmfkjgnolgimiaplp
- Read Aloud on Edge Add-ons: https://microsoftedge.microsoft.com/addons/detail/pnfonnnmfjnpfgagnklfaccicnnjcdkm
- Read Aloud privacy policy (Firefox Add-ons): https://addons.mozilla.org/en-US/firefox/addon/read-aloud/privacy/
- Read Aloud source: https://github.com/ken107/read-aloud